social media feeds becoming downstream of meme price discovery: when i ask onchain traders how they find alpha, the #1 response is tracking profitable wallets. in parallel, KOLs and influencers play a bigger role in price discovery across all markets than ever before..

Vibe coding and its consequences

cómo puede ser, presidente Javier Milei, que teniendo argentina una de las comunidades cripto más sólidas del mundo, nunca haya convocado a sus principales referentes antes de avanzar con regulaciones o acuerdos? cómo puede ser que vitalik buterin, probablemente la figura más

Someone asked about the basic statistics of the Sharpe Ratio. Here:

noticed that i no longer code that much. usually spend most of the coding session thinking and reviewing

our biggest release ever is now available to everyone in Cursor here's a rundown of what's new in 0.50 ↓

"Dive into Deep Learning" is a sprawling collection of code and theory (almost 1200pp) co-authored, among others, by Alex Smola. It is a very useful reference for specialized topics. arxiv.org/abs/2106.11342

1/ (re)introducing The Veldt: the social intelligence network where agents learn and users earn. personal ai agents monitor industry trends, build theses, and manage trades on your behalf. here’s why you’ll want one 👇

Acquiring knowledge is easy, the hard part is knowing what to apply and when. That’s why all true learning is “on the job.” Life is lived in the arena.

Cursor life hack. Start every new chat with the following prompt: "Do a deep-dive on the code and understand how [insert feature] works. Once you understand it, let me know, and I will provide the task I have for you." Reduces hallucinations by 10x

Don’t make things so you can make money; make money so you can make things.

have come to appreciate that agent engineering hinges on the harness—the optical cavity that turns chaotic photons into a laser and channels swarm intelligence into one coherent strike as model iq increases, the right harness makes the beam ever sharper

when you empower agents to articulate their own needs, the system design process transforms drastically the correct set up demands comprehensive observability—for both agent delegation and tool use—but once properly configured, building agent systems becomes a radically

always a good feeling when it is obvious what to do next

The problem with the crypto x AI space right now is that every project is out here shilling the same tired lines of removing gatekeepers, avoiding single points of failure, making things “decentralized” Basically, the same shit crypto has been repeating for years that no one

intelligence = ability to learn

Grok CLI >>> Cursor and Claude Code We wanted an IDE for xAI's Grok so we did something meta: we prompted the newly released Grok 4 to create... Grok CLI itself! Grok CLI can: 1. Modify local files and use the shell 2. Go through huge codebases and fix them 3. Persist for

very good model. jfc

How (and why) agents in Claude Code help you write better code, faster: Agents allow you to deploy specialized experts for every task. It's like managing a team rather than collaborating 1-on-1.